IT Is Not a Series of Checkboxes
-
I feel like it is becoming more and more common that people, both inside IT and in the management of companies that use IT, that IT is not a profession and does not require any special skills, planning or thought and that the industry is really just a club or a guild, if you will, where IT professionals maintain a secret list of checkboxes that, if checked, will result in the perfect IT infrastructure every time.
I can't blame managers and business people for this, although maybe they feel this way because HR departments and other heavily regulated functions nearly work this way at least when looked at by outsiders (e.g. do we have a valid insurance plan [yes/no], did he come into work late [yes/no]) but the blame really rests on the shoulders of IT "professionals" themselves. If you look at forums (not really this one, but many) you see people asking questions with zero business context, zero details wondering "what is the best SAN or storage device", "what's the best X or Y", etc. They think, quite often, that everything from hardware to operating systems to hypervisors to backup software to storage technologies to vendors to RAID levels to desktop wallpaper has a "best" or "always do this" option that if they just learn the secret they can skip knowing anything about IT and totally skip thinking about technology in the context of their specific business and just do the things on the checklist and they will "win" every time.
This is crazy. So many vendors, products, techniques and approaches exist because most are the "best" in different combinations, at different times for different scenarios. If there was a secret checklist that really had all the answers and required zero knowledge to use, it would have leaked long ago, be up on a website somewhere and all IT people in the world would be fired instantly and the entire field eliminated. That's not a risk that we are running.
IT is complex and requires a business context to create a technical context and both to be involved in every decision that is made. There are some checkboxes of what not to do, of course, as it is easy to make a solution that is truly horrible and useless or is deprecated because a universally better product or approach has arisen over time, but essentially nothing exists as an "always do this" or "always best" no matter what it is. If there was, all of its contenders would vanish, which I hope is obvious. (An example is RAID 2. This is a dead technology in every sense. But just because we know that RAID 2 should never be considered again doesn't tell us what RAID level to always use, we have many viable RAID levels to choose from for different situations. Yes, one can be eliminated but dozens remain - each is "best" in a different scenario. Some are commonly best, some are rarely best.) Even possibly the most likely to be "always true" rule in IT, always virtualize your server workloads has exceptions, rare ones and ones that are even more rare in the SMB, but they exist. Nothing is always, some things are never.
IT will never be so simple as to just look for a universal "best" or just go down a checklist. We always have to understand our business and our technology on an individual scenario level. We have to know, as IT decision makers, when to implement which solutions in which combinations. This is, truly, what being an IT decision maker is. If that went away, what would we even be doing?
-
@scottalanmiller said:
IT will never be so simple as to just look for a universal "best" or just go down a checklist. We always have to understand our business and our technology on an individual scenario level. We have to know, as IT decision makers, when to implement which solutions in which combinations.
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
-
@dafyre said:
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
A cost justification can often be explained from past examples where the current systems have failed, it simply needs to be explained correctly where the failure was, and how the failure could have been avoided.
SMB C-Level managers should be able to easily understand how to further avoid these failures if they only take the time to listen to the team/person they've hired.
-
@dafyre said:
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
This is one of my personal big issues - Once you've invested, it's really hard to let that investment go. Not only because you just spent that money/time/resources but because you might end up in this same exact spot after moving to the next product. You just paid this non-trivial amount to learn that this solution didn't work, and you might end up doing that again and again - puts some into a parallelized state.
This is compounded many fold when it's a major business app like an ERP or EHR (electronic health record). For example - there is no trialing an EHR. You basically have to look at someone else who is using it, ask all the questions you can, watch their workflow and guess - will it work for me? To setup a demo/test in your environment requires you to fully implement it (at least for a limited user base), and that is nearly as expensive as fully deploying it. -
@Dashrender said:
This is one of my personal big issues - Once you've invested, it's really hard to let that investment go. Not only because you just spent that money/time/resources but because you might end up in this same exact spot after moving to the next product. You just paid this non-trivial amount to learn that this solution didn't work, and you might end up doing that again and again - puts some into a parallelized state.
I totally agree here. Especially since any IT person worth their salt would know that No software is going to meet 100% of the needs. Once you are familiar with the Quirks and issues of your current setup, assuming there is no show-stopping reason to switch, it makes it that much easier to stay where you are at.
@Dashrender said:
This is compounded many fold when it's a major business app like an ERP or EHR (electronic health record). For example - there is no trialing an EHR. You basically have to look at someone else who is using it, ask all the questions you can, watch their workflow and guess - will it work for me?
This is also compounded by the fact that ERP / EHR implementations often run into the 6 digits or higher numbers... The only ERP migration I was a part of was $600k... (It was only supposed to have been around $300k, but overages, etc...)... A truly SMB probably can't afford to drop that kind of money on much of anything.
@Dashrender said:
To setup a demo/test in your environment requires you to fully implement it (at least for a limited user base), and that is nearly as expensive as fully deploying it.
And it also requires employees to do twice the work... (one entry in whatever the live system is, and one entry in the test system).
-
This is one of my personal big issues - Once you've invested, it's really hard to let that investment go.
A good example of this is Solidworks. For a long time they didn't have licensing for their software, so people shared it and never paid for it. Dassault finally started to license the software and the companies had no choice but to pay for it, because they would have more money in trying to learn another 3d CAD software as they would just paying for the license. It's not the best CAD software, but it's very widely used because people just were stuck with it.
I think another thing that trips up SMBs is they don't go to the right people with questions. I know the owner here got basically all his information from the reseller and didn't ask anyone else before moving to a new ERP.
-
@dafyre said:
@scottalanmiller said:
IT will never be so simple as to just look for a universal "best" or just go down a checklist. We always have to understand our business and our technology on an individual scenario level. We have to know, as IT decision makers, when to implement which solutions in which combinations.
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
Additionally, there is much in IT that cannot be observed (due to time, scale, cost, etc.) and so understanding the how and why is often necessary and extrapolating how things will be in the future.
-
@DustinB3403 said:
@dafyre said:
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
A cost justification can often be explained from past examples where the current systems have failed, it simply needs to be explained correctly where the failure was, and how the failure could have been avoided.
SMB C-Level managers should be able to easily understand how to further avoid these failures if they only take the time to listen to the team/person they've hired.
Only works if there has been time for that particular company to have experiences all failures in the past. That's a pretty horrible situation when good companies learn from the mistakes of others and avoid common mistakes or obvious mistakes up front either through following best practices, getting expert insight, applying common sense or dong research.
-
@Dashrender said:
@dafyre said:
Sometimes there is no clear-cut best. Sometimes it takes experimentation. A product demo isn't necessarily going to cover all of your bases. Even a 30 day demo in your own environment might not cover all of the bases. This is where a lot of the SMB market can struggle, I think. Once an SMB is invested in a certain way of doing things, then that way is difficult to change, especially when trying to cost-justify a change.
This is one of my personal big issues - Once you've invested, it's really hard to let that investment go. Not only because you just spent that money/time/resources but because you might end up in this same exact spot after moving to the next product. You just paid this non-trivial amount to learn that this solution didn't work, and you might end up doing that again and again - puts some into a parallelized state.
This is compounded many fold when it's a major business app like an ERP or EHR (electronic health record). For example - there is no trialing an EHR. You basically have to look at someone else who is using it, ask all the questions you can, watch their workflow and guess - will it work for me? To setup a demo/test in your environment requires you to fully implement it (at least for a limited user base), and that is nearly as expensive as fully deploying it.That's why companies love doing big demos like that..... oh, just run this in production for a month, if you are happy with it feel free to just keep it! The biggest cost to IT is often the implementation or at least that is a major cost and so once a working solution is in place people are very likely to keep it regardless of whether it is in any way the best solution.
-
@scottalanmiller said:
That's why companies love doing big demos like that..... oh, just run this in production for a month, if you are happy with it feel free to just keep it! The biggest cost to IT is often the implementation or at least that is a major cost and so once a working solution is in place people are very likely to keep it regardless of whether it is in any way the best solution.
Exactly! I'm not sure you can fix this though - without deploying/really using an ERP/EHR/etc you just don't if it will do what you want it to.
One of the issues I've seen around is not using the product as the product was intended to be used. I'm starting to move to the view that if you buy a boxed product, you pretty much need to use it as prescribed, or else you might be in for a world of hurt/disappointment.
-
@Dashrender said:
I'm starting to move to the view that if you buy a boxed product, you pretty much need to use it as prescribed, or else you might be in for a world of hurt/disappointment.
This is very true and a big deal that lots of people do not understand. Don't have someone else do all of the work then not leverage it; that's a bad combination.
-
@scottalanmiller said:
@Dashrender said:
I'm starting to move to the view that if you buy a boxed product, you pretty much need to use it as prescribed, or else you might be in for a world of hurt/disappointment.
This is very true and a big deal that lots of people do not understand. Don't have someone else do all of the work then not leverage it; that's a bad combination.
The problem is that many SMBs (and probably big companies at times too) don't want their processes to change.. they want the product to bend to them... This of course leads to frustrations and inefficiencies.
-
@Dashrender said:
The problem is that many SMBs (and probably big companies at times too) don't want their processes to change.. they want the product to bend to them... This of course leads to frustrations and inefficiencies.
Hence why they need a new article, lol.
-
@scottalanmiller said:
Hence why they need a new article, lol.
A new article? oh, as in one you're writing as we type?
-
@Dashrender said:
@scottalanmiller said:
Hence why they need a new article, lol.
A new article? oh, as in one you're writing as we type?
LOL, something like that.
-
@scottalanmiller said:
So many vendors, products, techniques and approaches exist because most are the "best" in different combinations, at different times for different scenarios.
Exactly. The best solution is the one that's most suited to a particular application/environment/budget. There never is a universal best - which, of course, is exactly why that perennial favourite of the interwebs - the "Top X Best" list - is pretty pointless.
-
@Brett-at-ioSafe said:
@scottalanmiller said:
So many vendors, products, techniques and approaches exist because most are the "best" in different combinations, at different times for different scenarios.
Exactly. The best solution is the one that's most suited to a particular application/environment/budget. There never is a universal best - which, of course, is exactly why that perennial favourite of the interwebs - the "Top X Best" list - is pretty pointless.
Those should just be called "Top X Most Popular" or "Top X That Paid Us To Review Them."
-
@johnhooks I thought that is what the Gartner's Magic Quadrant was?
-
@dafyre said:
@johnhooks I thought that is what the Gartner's Magic Quadrant was?
I had to look that up. I had never heard that before.
-
TBH, I'm not sure sure what it is either... I've just heard @scottalanmiller rail on them a time or two about being paid to give us good review type folks.
